HISTORY OF HEAVY OIL

Although the oil sands of Alberta have only been developed commercially since the late 1960s, their documented history dates back nearly three centuries to 1717, when Waupisoo of the Cree people brought samples of the oil sands to the Hudson’s Bay Company trading post at Fort Churchill (ERCB 2008b). Decades later, Peter Pond documented the oil sands at the confluence of the Clearwater River and the Athabasca, and in 1790 the oil sands region was visited and described by the European explorer, Sir Alexander Mackenzie. In 1875, a Geological Survey of Canada (GSC) expedition into the area allowed John Macoun, a botanist, to note that water washed the oil out of the oil sands. Seven years later, Robert Bell headed a second GSC survey in the area. Bell was the first to recognize that there was a potentially valuable petroleum resource in the area, and gave samples to a chemist named G. Christian Hoffman, who was successful in separating bitumen from the oil sands using water. Bell believed that there must be large reservoirs of oil deep under the ground, and that drilling would be required to extract it (Ferguson 1952). Based on Bell's view, 24 wells were drilled between 1906 and 1917 to locate the ‘pools’—with no success.During this period, experiments using the oil sands for road surfacing were conducted by Sidney Ells, an engineer with the mining branch of the GSC. Ells also found a company in California capable of extracting bitumen from sand using hot water (ERCB 2008), and conducted several experiments with the hot water flotation method.During and after World War 1 (1914-1918), Canada became increasingly aware that it was almost entirely dependent on foreign oil. This produced a sudden interest in discovering the nation’s oil resources. After World War 1, the Alberta Research Council (ARC) was formed by the provincial government to support oil sands research, among other projects. In the 1920s, an ARC scientist named Dr. Karl Clark developed a hot water flotation method that involved mixing oil sands with hot water and aerating the resulting slurry, which led to separation of bitumen froth from the sand. A field-scale oil sands separation plant, based on Dr. Clark’s design, was built near Fort McMurray in 1924. In 1928, Dr. Clark and his associate Sidney M. Blair were awarded a Canadian patent for the hot-water extraction process, a process still used today (Syncrude 2006).

The first commercial sale of bitumen from the oil sands occurred in 1930. Robert Fitzsimmons had constructed a small hot-water separation plant, based on the design of Dr. Karl Clark’s experimental plant, on the Bitumount site, and produced about 300 barrels of bitumen with a seven-man crew during the summer of 1930. After a series of owners and financial problems, the Bitumount site was taken over by the provincial government in 1948. While the site was eventually closed because the government was not interested in a commercial venture, data collected during the brief period of operation was used to evaluate the commercial viability of the oil sands. In 1950, the provincial government announced that it was indeed a viable undertaking.The beginning of modern-day commercial oil sands development began in 1953, when the Great Canadian Oil Sands consortium—which would become Suncor Inc. in 1979—was formed. Construction of the Great Canadian Oil sands plant began in 1964, and production began in 1967.The Syncrudeconsortium was formed in 1964, with an initial objective of researching the economic and technical feasibility of mining oil from the Athabasca oil sands (Syncrude 2006). Construction of the Mildred Lake facility began in 1973, and the first barrel was shipped in 1978.Imperial Oil began production at the first commercial in situproject, in Cold Lake, in 1985, with production exceeding 140,000 barrels per day by 1989. Before that time, natural gas liquids and their byproducts dominated production; Imperial Oil was largely responsible for increasing the production of bitumen five-fold from 1984 to 1996.Other companies or consortiums that have joined the Alberta oil sands boom, and that are now members of RAMP, include:Albian Sands Energy Inc., Shell Canada Limited, Canadian Natural Resources Limited, Petro - Canada Oil and Gas ,OPTI Canada Inc., Nexen Inc., Husky Energy, Total E&P Canada Ltd., and Birch Mountain Resources Ltd. Through projects developed by these and other companies, bitumen production from the oil sands reached 1.255 million bpd in 2006 (AEUB 2007).

INTRODUCTION

Heavy oil is generally defined using API gravity and may also include viscosity in the definition. API gravity was established as a uniform way of characterizing the density or specific gravity of oil by the American Petroleum Institute. API gravity is an arbitrary scale expressing the gravity or density of liquid petroleum products. The measuring scale is calibrated in terms of degrees API. It is calculated as follows:

API Gravity (°) = (141.5 ÷ specific gravity of the oil at 60ºF) − 131.5

Higher API gravity ratings reflect lighter types of crude oil. The boundaries between different classes of oil (e.g., light, intermediate, heavy, extra heavy) all follow the same trend, but different authors choose slightly different boundaries between categories. Several examples are listed below.

DOE’s Energy Information Administration (EIA) Petroleum Navigator tool on the EIA website 1 offers the following definitions:

• Light crude has a gravity of greater than 38° API.

• Intermediate crude ranges from 22°–38° API.

• Heavy crude has a gravity of less than 22° API.

The U.S. Geological Survey (USGS) also considers the viscosity of the oil and provides the following definitions in Meyer and Attanasi (2003).

• “Light oil, also called conventional oil, has an API gravity of at least 22° and a viscosity less than 100 centipoise (cP). • Heavy oil is an asphaltic, dense (low API gravity), and viscous oil that is chemically characterized by its content of asphaltenes (very large molecules incorporating most of the sulfur and perhaps 90 percent of the metals in the oil). Although variously defined, the upper limit for heavy oil has been set at 22° API gravity and a viscosity of 100 cP.

• Extra-heavy oil is that portion of heavy oil having an API gravity of less than 10°.

• Natural bitumen, also called tar sands or oil sands, shares the attributes of heavy oil but is yet more dense and viscous. Natural bitumen is oil having a viscosity greater than 10,000 cP.” According to the Canadian Centre for Energy Information, 2 the Canadian industry defines terms as follows:

• Light crude oil has API gravity higher than 31.1°.

• Medium oil has API gravity between 31.1° and 22.3°.

• Heavy oil has API gravity between 22.3° and 10°.

• Extra heavy oil (bitumen) has API gravity of less than 10°.

The Canadian Centre for Energy Information also notes that the Canadian government has only two classifications:

• Light oil has API gravity of greater than 25.7°.

• Heavy oil has API gravity of less than 25.7°.

Dusseault (2001) recommends that viscosity be measured in situ, and that heavy oil has viscosity greater than 100 cP. He further suggests that the definition for heavy oil could also be expressed in terms of produceability. Heavy oil should have some mobility under naturally existing conditions and can flow to wells and be produced economically. In contrast, extra heavy oils, oil sands, and bitumen typically have both low API gravity and high viscosity, such that they do not flow naturally. They are typically produced through thermal processes or solvent addition. World conventional oil (“light oil”, >20°API) production from natural sources must eventually peak and enter into decline because of increasing

world demand, inexorable reservoir production rate decline, a fixed resource base, and the indisputable fact that few new sedimentary basinsremain to be exploited. Many believe this will occur in the period 2005-2010. Current consumption is approximately 77 MBOD (Oil and Gas Journal), Canada produces about 2.2 MBOD. This is about 3.3% of the world total, but a disproportionate amount of this figure comes from <20°API heavy oil, more than of 45% of Canadian oil production, or about 1.5% of the world’s total oil production. Of world production, about 7-8 % of the total comes from heavy oil.After the peak in world conventional oil production rate is passed, perhaps about 5 years from now, light oil production will gradually decline at a rate that will be somewhat tempered, but not reversed by the gradual introduction of, among others, the new Canadian-developed technologies such as gravity drainage and pressure pulsing. The advent of superior 3-D seismic methods and geochemistry analysis in the last 20 years, combined with the knowledge generated from the huge database that is now extant for hundreds of basins world-wide, mean that when a new basin is explored, the potential resource capacity of the basin can be well-bounded, even if only a few exploratory wells have been drilled. Essentially, all land-based basins have been so explored, and few remote basins remain totally unevaluated. For example, the large Falkland Islands (Malvina Islands) Basin offshore Argentina has been explored in the last few years, and it is one of the few remaining large basins that remained untouched. Incidentally, it has turned out to be disappointing, based on preliminary exploratory drilling by Phillips and Shell in the period 1997-2000.Simply put, the world is running out of conventional oil because the world is running out of new basins to explore and exploit. Furthermore, the remaining basins are remote, and remote resources are quantity and cost-limited: exploitation costs are high in deep and remote basins(deep offshore, Canadian Pacific Coast, Antarctic fringe, Arctic basins…), therefore only larger finds will be developed, and recovery ratios will be less than for “easy” basins. It is unlikely that these basins will provide any more than another 10-15% of the conventional oil already known to be in place in explored basins.

Types and Significant Deposits of Heavy Oil

The two main forms of heavy oil typically described in the literature are viscous heavy oil and oil sands (bitumen). While some examples of each are clearly distinct, there is a gradient in properties that blurs the boundary between viscous crude found in a sandstone formation and oil sands. Several examples of each are described below.

Viscous Heavy OilThe first type of heavy oil described here is liquid or semi-liquid but is very viscous. In many parts of the world, heavy oil seeps to the surface and accumulates in pits or other depressions.

CaliforniaWhen Spanish explorers landed in California in the 1500s, they found Indians using asphaltum (very thick oil gathered from natural seeps) to make baskets and jars, to fasten arrowpoints to shafts, and for ornaments. The explorers, in turn, used asphaltum to seal seams in their ships (Ritzius et al. 1993). The history of oil development in California is documented in Ritzius et al. (1993) and through an interesting website of the San Joaquin Geological Society. 3 Another well known example of natural accumulations of viscous heavy oil in California is the La Brea tar pits located near Los Angeles.California proved to have abundant oil reserves. By the late 1800s, oil was being produced through drilled wells. Exploration throughout the state found at least six giant oil fields, three of which contain heavy oil. The Midway-Sunset, Kern River, and South Belridge fields have produced more than 1 billion barrels of oil each (Curtis et al. 2002). DOE’s EIA reports that California produced nearly 217 million bbl of crude oil in 2007. 4 The EIA website does not differentiate between heavy oil and other forms of oil.The Kern River field began production prior to 1900 and continues today (Figure 1). Curtis et al. (2002) report that the Kern River field has an API gravity of 10° to 15° and a viscosity of 500 to 10,000 cP. These features, along with the low initial reservoir temperature and pressure, led to a modest primary recovery. In the 1960s, the industry began trying steam injection to help the heavy oil flow more readily. Kern River crude oil

reacted well to steam flooding, and the production rates increased substantially.

VenezuelaVenezuela is home to several large heavy oil fields (Figure 2). The western part of the country, around Lake Maracaibo, holds large reserves of heavy oil. The API gravity of the crude oil in the Maracaibo region ranges from 9° to 33° (Dusseault 2008a).But the largest accumulation of extra heavy oil in the world is found in a zone in central Venezuela known as the Faja Petrolifera del Orinoco (often shortened to Faja del Orinoco or just Faja). Dusseault et al. (2008) note that the Faja is estimated to hold almost 1.3 trillion barrels of oil in place. The extra heavy crude oil here has a typical API gravity of 7° to 10°. However, unlike many other low API-gravity crudes, the viscosity of Faja crude is somewhat lower, thereby allowing the crude to be partially produced without thermal techniques. Later technology advances have allowed greater production of the Faja.

Figure 1. Map of Kern River Oil Field in California

Figure 2. Map of Venezuela showing major heavy oil fields

Conspicuously absent from most of the heavy oil literature is oil shale. Oil shale in its natural state contains kerogen, a precursor to petroleum. Kerogen is the solid, insoluble, organic material in the shale that can be converted to oil and other petroleum products by pyrolysis and distillation. The kerogen in oil shale does not flow naturally and must be subjected to heat treatment to be released from the shale. Nevertheless, the world will never run out of oil, for several reasons. First, conventional oil comprises a small fraction of hydrocarbons in sedimentary basins (Table 1). As price pressure increases in the future, and as now extraction methods and processing technologies are developed and perfected (such as coal conversion and shale oil extraction), these resources will become available. The time frame for these, including the methane hydrate resource in the deep ocean, is probably several generations.

Second, as technology evolves, other energy sources (hydrogen cycle?) will displace oil just as oil displaced coal. Even now, natural gas is displacing oil in fleet vehicles in many cities (Toronto, Los Angeles, Beijing), and this trend will continue. However, natural gas is a valuable resource, and it too is limited. Alberta’s production rate of natural gas is expected to peak in the next two years, and for the entire world, the peak should arrive in about 15-18 years. Third, there are new technologies emerging for greater energy efficiency and energy recycling.For example, in 2002, the City of Los Angeles will likely start disposing of municipal biosolids through deep injection. At the high temperatures at depth (50-80°C), anaerobic methanogenic bacteria can degrade all the free hydrogen in the carbohydrate-rich organic wastes into CH4, within several years, and if the wastes are placed into a suitable geological formation, the evolved gas can be collected and re-used. This emerging technology is suitable not only for municipal biosolids, but can be used for any organic material (animal wastes from feed lots,sawdust, etc.). Not only will this technology generate CH4, it seems to be environmentally benign and cost competitive with

current waste treatment methods, even without factoring in the value of the gas that is generated.

World and Canadian Heavy Oil Resources This chat below show the distribution of oil below

The world resource is about 12×1012 bbl,or about 2000 cubic kilometres of volume, a cube with sides 12.6 km long. Apparently, there is more than twice as much resource available in <20°API oil as in conventional oil >20°API.Furthermore, it is widely believed that the heavy oil resource is somewhat underestimated in comparison to the conventional oil resource because of poorer-quality data. Note that this amount of heavy oil does not include shale oil, a resource that is vaster, but much lessconcentrated in terms of unit volume, and also much less accessible because of inherent low permeability. Heavy oil resources are found

throughout the world, but Canada and Venezuela are singularly endowed. The two countries appear to share 35-40% of the world resources of <20°API heavy oil, approximately 2.5×1012 bbl in the Canadian HOB and Oil Sands regions, and 1.5×1012 bbl in the Venezuelan Faja del Orinoco tar sands belt (based on recently published estimates in the 2001 Margarita Conference). The specific amount, particularly in Venezuela, is to a degree conjectural: it depends on the definition of what is an oil stratum in terms of thickness and oil saturation level. Nevertheless, given that similar uncertainties exist for such estimates around the world (the oil industry in many regions is only beginning to take an interest in heavy oil), the actual amounts quoted here are reasonable estimates. Other countries with appreciable heavy oil resources include Russia, Nigeria, Indonesia and China, as well as several of the Middle East nations (well-endowed with conventional) where more shallow heavy oil has been ignored because of the large production capacity of their conventional oil reservoirs. To put the available heavy oil resource into an understandable context, its size in Canada alone is so large (~350-400×109 m3, more than 20% of the World total) that, at a stable combined US and Canadian consumption rate of ~1.2×109 m3/yr, there is enough heavy oil in Canada to meet 100% of this demand for over 100 years if the overall extraction efficiency is ~30%. In the best strata, the new extraction technologies in Canada are already approaching, and in some cases exceeding, this recovery ratio of 30%. Oil sands mines approach 85% extraction. Imperial Oil Limited at Cold Lake is approaching 25% extraction. The four new SAGD projects initiated inCanada in 2001 (Foster Creek – AEC; MacKay River – Petro-Canada; Surmont – Gulf Canada, now Conoco;Primrose – CNRL) expect to achieve >50% extraction. In good reservoirs, CHOPS can exceed 20% recovery.An informal poll of Canadian petroleum engineers and scientists involved in new production technologies in 2001 yielded the following estimates of ultimate recovery of Canadian heavy oil (including present, emerging, and yet-to-be developed technologies):1. 20% extraction with 95% certainty2. 35% extraction with 50% certainty3. 50% extraction with 5% certainty Evidently, even given the persistent optimism of engineers, the expectations for reasonable recovery ratios are clear: a large percent of the heavy oil is currently economically and technologically available, and more

will become so. Using a factor of 35%, and assuming (optimistically) that the recovery ratio from conventional oil will eventually reach 60%, there is still more heavy oil available as a future resource than all the conventional oil that has been or will be produced. Canadian Heavy Oil Belt Resources

Canada is blessed with huge heavy oil resources in Alberta and Saskatchewan (Figure 3). The northern deposits are true tar sands (or oil sands) with combinations of extra-heavy crude oil and bitumen (<10°API) of high viscosity (>50,000 cP in situ) filling the sandstone interstices. These are discussed in a later section. Other deposits can be considered as viscous heavy oil and are therefore mentioned here.

Figure 3. Location of Canadian oil sands and viscous heavy oil deposits

The more southerly and easterly deposits make up a large region of heavy oil deposits (known as the Heavy Oil Belt), found in a series of blanket sands and channel sands extending all the way from southwest Saskatchewan to zones overlying the Cold Lake Oil Sands near Bonnyville, (located about 120 km north of Lloydminster). The oil is considerably lighter in density (11° to 18° API gravity) and of much lower viscosity (500 to 20,000 cP) as compared to the major oil sands deposits to the north, therefore it is easier to produce, which is why it is the focus of much of the recent increases in heavy oil production. There are perhaps 300 billion barrels of oil in place in the Heavy Oil Belt, and it is estimated that at least 50–60 billion barrels may ultimately be recoverable (Dusseault 2001).The recent NEB report on Conventional Heavy Oil resources of the Western Canada Sedimentary Basin (2001, see footnote 1) has identified 50×109 m3 (~350×109 bbl) of heavy oil in place in the HOB. This is about 15% of the total <20°API resource in Alberta, exclusive of the ill-defined Carbonate Triangle. They estimate that 21% of this, ~74×109 bbl, can be recovered with current technology. This is 1000 days of supply for the entire world at current consumption rates. Given Canada’s light population, it is of huge economic importance, with a current commodity value somewhat below CAN$3×1012 at a world price of US$25.00. Given the technological progress that is ongoing, referring to the 50% probability estimate of Canadian industry engineers, the NEB estimate of technologically accessible reserves in the HOB is conservative, perhaps by a factor of two. The writer believes that a reasonable estimate of the recovery from the HOB in Canada is on the order of 150×109 bbl. Importantly, based on interaction with many producers, the NEB published estimates of the amount of economically produceable oil. If operators can accept total costs of ~CAN$13.00/bbl, ~80% of the HOB resource is economically accessible. This can be compared to the average price of about CAN$18.00 – 25.00 that producers received in the period Jan-Sept 2001 (bitumen at Cold Lake from $9.62 to $28.80/bbl in this period, and Lloyd blend at Hardisty from $22.00 to 34.00/bbl). Apparently, primary heavy oil production through CHOPS methods is currently quite profitable, despite the historically high differential price between conventional feedstock and heavy oil.

Heavy Oil Production Technologies

Heavy oil deposits are found in many parts of the world in many different geological and climatic settings. These factors, along with the viscosity and API gravity of different heavy oil deposits, lead to a wide array of technologies for producing the oil. The technologies differ in several important ways:

• Mining vs. in situ processes,

• Cold (ambient temperature) vs. thermal processes,

• Technologies already in common use vs. emerging technologies.

The following sections include brief descriptions of the key technologies. Heavy viscous oil and oil sands share comparable technologies; these are described together. Some of the technologies employed to produce oil shale follow different processes; therefore, oil shale is included in a separate section.

CHOPS Cold Heavy Oil Production with Sand (CHOPS) is a production technique that operates contrary to the conventional wisdom that sand should be blocked from entering a well. Perhaps the most thorough discussion of CHOPS technology is found in Dusseault (2001). Readers are encouraged to consult that reference for much more detailed information on CHOPS.CHOPS technology encourages production of sand from unconsolidated sandstone reservoirs. As the produced sand moves from the formation into the well, it leaves behind channels referred to as “wormholes.” This increases permeability near the wellbore and allows more oil to reach the wellbore. Heavy oil production has increased 10- to 20-fold after converting wells from traditional production to CHOPS production (Hart Energy 2006).CHOPS technology typically uses vertical wells fitted with PCPs to move the large volume of sand to the surface (Figure 8). PCPs are more effective for pumping the sand-laden material to the surface. PCPs typically include a stainless steel rotor mechanism that moves inside of an elastomer-lined helical cavity. Sand production initially can be as high as

40% by volume of the produced material. Later, the sand concentration drops but still remains high at 0.5% to 10% by volume (Dusseault 2001).Dusseault (2008a) provides many examples of the dramatic increase in heavy oil production from individual wells in the Luseland and Edam fields in Saskatchewan that had been previously operated for primary production. For an example, see Figure 9. When the wells were converted to CHOPS configuration and operation, the oil production increased dramatically and resulted in large incremental production over the life of the well. Dusseault also notes that there are hundreds of CHOPS fields in the Heavy Oil Belt of Canada. In 2003, those wells contributed about 700,000 bbl/day of oil production. The oil had a viscosity range of 50–15,000 cP (most fields are >1,000 cP). Wells were completed at depths from 360 to 900 m (Dusseault 2008a). Collins et al. (2008) report on the use of CHOPS in the Karazhanbas Field, a giant shallow heavy oil field in western Kazakhstan. The heavy oil deposit is less than 460 m deep and contains heavy oil (~400 cP) in seven reservoir zones. PCPs are used to lift the oil, and sand is allowed to enter into perforated zones. Production of about 38,000 bbl/day was reached by January 2004, an increase of over 25,000 bbl/day within 4 years. Sand flux is far lower than in Canadian cases because of low oil viscosities.

The major heavy oil production technology discussed will be CHOPS: Cold Heavy Oil Production with Sand. CHOPS involves deliberate initiation of sand influx into a perforated oil well, and continued production of substantial quantities of sand along with the oil, perhaps for many years. CHOPS requires management of large quantities of sand in all phases of production; this is a radically different concept to conventional oil well production management. Also, there are physical processes occuring in the reservoir that are completely foreign to conventional oil production engineers (foamy oil behavior, massive stress redistribution, liquefaction of sand, flow of a four-phase slurry…). Because CHOPS requires a radically different approach to oil field management and because scientific and engineering personnel have to learn new physical principals and apply them, CHOPS qualifies as a new oil production technology. It is a primary production method because it exploits natural energy sources in the reservoir: energy from dissolution and expansion of gas (compressional energy), and energy from the downward motion of the overburden (gravitational energy). It is now widely understood in heavy oil exploitation that the exclusion of sand during primary production through use of screens or gravel packs in vertical wells3 means that the oil cannot be produced economically. Individual vertical well rates will be only a few cubic metres per day (0.5 – 5 m3/d), and the best of these rates will be attained only in the lower viscosity heavy oils (<1000 cP) and the better reservoirs (k > 2 D, t > 10 m). If sand ingress is initiated and sustained in reservoirs that have the right characteristics, oil production rates as high as 15-50 m3/day canbe achieved in almost all cases. Such rates have also been achieved without large-scale sand influx in some heavy oil reservoirs exploited with long horizontal wells. However, the cost of a horizontal well is and will remain about 3× to 5× more expensive than a vertical well, and well intervention costs are extremely high in horizontal wells. These horizontal wells usually have a life-span of 3-6 years only, and typically no more than 10% of the original oil in place (OOIP) is produced. Data will also be presented later to show that in most cases CHOPS wells are more profitable and produce more total oil than horizontal wells. CHOPS produces large quantities of oily sand as well as various categories of fluid wastes: chloride-rich water (dissolved NaCl), water-oil-clay emulsions, slops, tank bottom sludges, and soil-fluid mixes arising from spills. The handling of all these wastes, including the massive volumes of produced

sand, can add as much as $3.00/bbl to the operating costs (OPEX) for CHOPS. Waste management is considered to be the major cost factor for CHOPS operations,resulting in about 15-35% of OPEX, depending on oil and sand rates. Understanding and minimizing these costs are fundamental to planning and executing a heavy oil project using CHOPS methods. The other large OPEX costs category is the cost of well workovers. CHOPS wells need far more frequent workovers than conventional oil wells, and this results in about 15-25% of OPEX, depending on the field and the wells. Reducing not only the fraction of OPEX absorbed by workovers, but also reducing all OPEX costs from the current level of about $7.00/bbl could increase profits, open up more marginal fields for development, and also allow the redevelopment of currently inactive (but not abandoned) wells. Before 1985-1990, heavy oil production was based largely on thermal stimulation ( T – changes in temperature) to reduce viscosity and large Δpressure drops ( p – changes in pressure) to induce flow. Projects used Δcyclic steam stimulation (huff-‘n-puff), steam flooding, wet or drycombustion using air or oxygen injection, and combinations of these methods. Until recently, these technologies employed arrays of vertical to mildly deviated wells (<45°). Only three projects in Canada still use these old techniques. Some methods have never proven viable for heavy oil: these include solvent injection, biological methods, cold gas (CH4, CO2…) injection, polymer methods, and in situ emulsification. Up to about 1985-88, marginally economical non-thermal production with vertical wells was used in a limited manner in Alberta and Saskatchewan, but these wells produced less than 10 m3/d, recovery was invariably less than 5-8% OOIP (Original Oil In Place), and small amounts of sand generally entered the wellbore during production. Note that all high-pressure methods experience advective instabilities such as viscous fingering, permeability channeling, water or gas coning, and uncontrolled (upward) hydraulic fracture propagation. These instabilities result in bypassing oil, isolating bodies of the oil by sweeping permeable channels clear of oil, early loss of wells because of excessive water production or gas production, early loss of reservoir energy, and so on.

Slurrified heavy oil recovery process

In at least one specific embodiment, a method for recovering heavy oil includes accessing, from two or more locations, a subsurface formation having an overburden stress disposed thereon, the formation comprising heavy oil and one or more solids. The formation is pressurized to a pressure sufficient to relieve the overburden stress. A differential pressure is created between the two or more locations to provide one or more high pressure locations and one or more low pressure locations. The differential pressure is varied within the formation between the one or more high pressure locations and the one or more low pressure locations to mobilize at least a portion of the solids and a portion of the heavy oil in the formation. The mobilized solids and heavy oil then flow toward the one or more low pressure locations to provide a slurry comprising heavy oil and one or more solids. The slurry comprising the heavy oil and solids is flowed to the surface where the heavy oil is recovered from the one or more solids. The one or more solids are recycled to the formation.Embodiments of the invention relate to in-situ recovery methods for heavy oils. More particularly, embodiments of the invention relate to water injection methods for heavy oil recovery from sand and clay. Bitumen is a highly viscous hydrocarbon found in porous subsurface geologic formations. Bitumen is often entrained in sand, clay, or other porous solids and is resistant to flow at subsurface temperatures and pressures. Current recovery methods inject heat or viscosity reducing solvents to reduce the viscosity of the oil and allow it to flow through the subsurface formations and to the surface through boreholes or wellbores. Other methods breakup the sand matrix in which the heavy oil is entrained by water injection to produce the formation sand with the oil; however, the recovery of bitumen using water injection techniques is limited to the area proximal the bore hole. These methods generally have low recovery ratios and are expensive to operate and maintain.In another approach, the method described in commonly assign utilizes separate bore holes for water injection and production. That method first relieves the overburden stress on the formation through water injection and then causes the hydrocarbon-bearing formation to flow from the injection bore hole to the production bore hole from which the heavy oil, water, and formation sand is produced to the surface. Once the heavy oil is removed from the formation sand, the hydrocarbon-free sand is reinjected with water to fill the void left by the producing the slurry. Although the '631 method is a significant step-out

improvement over conventional water injection techniques, there is still a need for further improved methods for continuously and cost-effectively recovering bitumen from subsurface formations.Embodiments of the present invention provide improved methods for continuously and cost-effectively recovering heavy oils from subsurface formations.In at least one specific embodiment, the method includes accessing a subsurface formation having an overburden stress disposed thereon from two or more locations, the formation comprising heavy oil and one or more solids. The formation is pressurized to a pressure sufficient to relieve the overburden stress. A differential pressure is created between the two or more locations to provide one or more high pressure locations and one or more low pressure locations. The differential pressure is varied within the formation between the one or more high pressure locations and the one or more low pressure locations to mobilize at least a portion of the solids and a portion of the heavy oil in the formation. The mobilized solids and heavy oil then flow toward the one or more low pressure locations to provide a slurry comprising heavy oil and one or more solids. The slurry comprising the heavy oil and solids is flowed to the surface where the heavy oil is recovered from the one or more solids. The one or more solids are recycled to the formation.In at least one other specific embodiment, the method includes accessing, from two or more locations, a subsurface formation having an overburden stress disposed thereon, the formation comprising two or more hydrocarbon-bearing zones containing heavy oil and one or more solids; injecting a fluid into the formation at two or more depths within the formation and pressurizing at least one of the two or more hydrocarbon-bearing zones within the formation to a pressure sufficient to relieve the overburden stress; causing a differential pressure within the formation to provide one or more high pressure locations and one or more low pressure locations within the at least one of the two or more hydrocarbon-bearing zones within the formation; varying the differential pressure within the formation to mobilize at least a portion of the heavy oil and a portion of the one or more solids; causing the mobilized one or more solids and heavy oil to flow toward the one or more low pressure locations to provide a slurry comprising heavy oil and one or more solids; flowing the slurry comprising the heavy oil and one or more solids to the surface and recovering heavy oil from the slurry comprising heavy oil and one or more solids. Then recycling the one or more solids to the formation.In yet

another specific embodiment, the method includes accessing, from two or more locations, a subsurface formation having an overburden stress disposed thereon, the formation comprising two or more hydrocarbon-bearing zones containing heavy oil and one or more solids; injecting a fluid into the formation at two or more depths within the formation;pressurizing at least one of the two or more hydrocarbon-bearing zones within the formation to a pressure sufficient to relieve the overburden stress; causing a differential pressure within the formation to provide one or more high pressure locations and one or more low pressure locations within the at least one of the two or more hydrocarbon-bearing zones within the formation; varying the differential pressure within the formation to mobilize at least a portion of the heavy oil and a portion of the one or more solids, thereby providing mobilized one or more solids and heavy oil; causing the mobilized one or more solids and heavy oil to flow toward the one or more low pressure locations to provide a slurry comprising heavy oil and one or more solids; flowing the slurry comprising the heavy oil and one or more solids to the surface; recovering heavy oil from the slurry comprising heavy oil and one or more solids; and recycling the one or more solids to the formation. A detailed description will now be provided. Each of the appended claims defines a separate invention, which for infringement purposes is recognized as including equivalents to the various elements or limitations specified in the claims. Depending on the context, all references below to the “invention” may in some cases refer to certain specific embodiments only. In other cases it will be recognized that references to the “invention” will refer to subject matter recited in one or more, but not necessarily all, of the claims. Each of the inventions will now be described in greater detail below, including specific embodiments, versions and examples, but the inventions are not limited to these embodiments, versions or examples, which are included to enable a person having ordinary skill in the art to make and use the inventions, when the information in this patent is combined with available information and technology.

FIG. 1 is a schematic diagram of a multi-wellbore system 100 for producing heavy oil from a subsurface formation according to one or more embodiments described. The multi-wellbore system 100 can include two or more wellbores 110, 120 (only two shown). Each wellbore 110, 120

extends from the surface through the overburden 130 and accesses a formation 140 that includes one or more hydrocarbon-bearing zones 145 (only one shown) from which heavy oil is to be produced and recovered.The term “heavy oil” refers to any hydrocarbon or various mixtures of hydrocarbons that occur naturally, including bitumen and tar. In one or more embodiments, a heavy oil has a viscosity of at least 500 cP. In one or more embodiments, a heavy oil has a viscosity of about 1000 cP or more, 10,000 cP or more, 100,000 cP or more, or 1,000,000 cP or more.The term “formation” refers to a body of rock or other subsurface solids that is sufficiently distinctive and continuous that it can be mapped. A “formation” can be a body of rock of predominantly one type or a combination of types. A formation can contain one of more hydrocarbon-bearing zones.The term “hydrocarbon-bearing zone” refers to a group or member of a formation that contains some amount of heavy oil. A hydrocarbon-bearing zone can be separated from other hydrocarbon-bearing zones by zones of lower permeability such as mudstones, shales, or shaley sands. In one or more embodiments, a hydrocarbon-bearing zone includes heavy oil in addition to sand, clay, or other porous solids.The term “overburden” refers to the sediments or earth materials overlying the formation containing one or more hydrocarbon-bearing zones. The term “overburden stress” refers to the load per unit area or stress overlying an area or point of interest in the subsurface from the weight of the overlying sediments and fluids. In one or more embodiments, the “overburden stress” is the load per unit area or stress overlying the hydrocarbon-bearing zone that is being conditioned and/or produced according to the embodiments described.The term “wellbore” is interchangeable with “borehole” and refers to a man-made space or hole that extends beneath the surface. The hole can be both vertical and horizontal, and can be cased or uncased. In one or more embodiments, a wellbore can have at least one portion that is cased (i.e. lined) and at least one portion that is uncased.Referring to FIG. 1, an injection fluid is introduced to the hydrocarbon-bearing zone 145 through a first wellbore 110 (“injection wellbore”) via stream 150. A production slurry exits the hydrocarbon-bearing zone 145 and is conveyed (“produced”) through a second wellbore 120 (“production wellbore”) via stream 160. The production slurry can include any combination (i.e. mixture) of heavy oil, clay, sand, water, and brine. The production slurry can be transferred via

stream 160 to a recovery unit 170 where the heavy oil is separated and recovered from the solids and water. The recovery unit 170 can utilize any process for separating the heavy oil from the solids and water. Illustrative processes include cold water, hot water, and naphtha treatment processes, for example.The recovered heavy oil (with possibly some residual solids and water) from the recovery unit 170 is then passed via stream 180 for further separation and refining using methods and techniques known in the art. The hydrocarbon-free or nearly hydrocarbon-free solids and recovered water from the recovery unit 170 can be recycled to the injection wellbore 110 via recycle stream 190, as shown in FIG. 1. The solids, water, or mixture of the solids and water can then be re-injected into the formation 140 via stream 150. Depending on process requirements, additional water or solids can be added to the recycle stream 190 or water or solids can be removed from the recycle stream 190 to adjust the solids concentration of stream 150 prior to injection through the wellbore 110 to the formation 140. Other fluids or solids including fresh sand or clay can also be added to the recycle stream 190 as needed. Conditioning Phase In operation, the injection fluid is pumped or otherwise conveyed through the injection wellbore 110 via stream 150 into the hydrocarbon-bearing zone 145 of the formation 140. One purpose of the injection fluid is to raise the fluid pressure in the formation 140 and relieve the overburden stress on the formation 140 (i.e. to “condition” the formation). Accordingly, the pressure of the injection fluid should be sufficient to relieve the overburden 130. Another purpose of the injection fluid is to increase the initial porosity of the formation 140 and therefore, increase the permeability of the formation 140 to the injected fluid (generally water or brine) as well as to partially or totally break up or disaggregate (through shear dilation) a portion of the shale or mudstone layers that may be embedded within the hydrocarbon-bearing zones 145 of the formation 140. This could remove those shale or mudstone layers from acting as baffles or barriers to the fluid flow within the formation 140 between the injection wellbore 110 and production wellbore 120.Therefore, the pressure of the injection fluid should also be sufficient to permeate through the hydrocarbon-bearing zone 145 and develop a relatively constant pressure within the hydrocarbon-bearing zone 145 of the formation 140 at

the end of conditioning. Preferably, the pressure of the injection fluid is at or above the stress of the overburden 130 exerted on the hydrocarbon-bearing zone 145 to allow the formation of horizontal or sub-horizontal fractures in the hydrocarbon-bearing zone. When the stress of the overburden 130 is relieved or nearly relieved throughout a majority of the volume of the hydrocarbon-bearing zone from which heavy oil production is planned, the hydrocarbon-bearing zone 145 is considered to be “conditioned.”FIG. 2 is a schematic illustration of an alternative embodiment of the multi-wellbore system 100 of FIG. 1 where injection fluid is passed through both wellbores 110 and 120 for conditioning the formation 140. The injection fluid can be injected into the hydrocarbon-bearing zone 145 through both the injection wellbore 110 and the production wellbore 120 to substantially reduce the time required to equalize the stress of the overburden 130, as shown in FIG. 2. For example, the time to relieve the stress of the overburden 130 can be reduced by as much as half or more.Further more, the injection fluid can be injected into the hydrocarbon-bearing zone 145 through both the injection wellbore 110 and the production wellbore 120 to break or disaggregate (through shear dilation) a greater portion of the shale or mudstone layers that may be dispersed within the hydrocarbon-bearing zones 145 of the formation 140. At the very end of the conditioning process, the injection of fluid at a high rate through the production wellbore 120 can also help the early onset of slurry production through the production wellbore 120 by breaking up any near wellbore shale or lithified rock fragments that may impede the uniform displacement of the hydrocarbon-bearing zone 145 and slurrifying the solids immediately adjacent to the wellbore.Furthermore, the injection fluid can be emitted either simultaneously or sequentially through both wellbores 110, 120 as shown in FIG. 3 to create or cause fractures to propagate from near each wellbore 110, 120 into the formation, thereby allowing the injected fluid greater access to the formation and increasing the porosity/permeability throughout a greater area and/or volume within the hydrocarbon-bearing zone 145 more quickly. By introducing injection fluid from multiple locations within the same formation 140, the hydraulically-induced horizontal (or sub-horizontal) fractures and/or natural flow conduits 305 can help access and contact a larger portion of the formation 140 with fluid than could be from the drilled wellbore alone.

In addition, by injection at multiple depths within the formation and creating horizontal (or sub-horizontal) fractures at those multiple depths, the distance the injected fluid has to flow to pressurize or condition the reservoir is greatly reduced. In areas where hydraulically induced fractures may propagate in directions such that they do not contact a sufficient volume of the hydrocarbon-bearing zone, man-made or natural conduits to fluid flow may aid in accelerating the dispersement of injected fluid and pressure throughout the hydrocarbon-bearing zone. These man-made conduits could include horizontal wells, channels or wormholes created from previous fluid and solids production or natural zones of higher absolute permeability or higher water saturation (and therefore higher permeability to the injected water). As mentioned, the injection fluid can dilate, break, or otherwise disaggregate at least a portion of the shale or mudstone layers 310 that are embedded within the hydrocarbon-bearing zone 145 of the formation 140 thereby increasing the permeability of these materials to the injected fluids. If not broken or dilated, such shale or mudstone layers can act as baffles or barriers that impede the flow of the injected fluids through the hydrocarbon-bearing zone 145. Furthermore, the injection fluid can more quickly distribute throughout the hydrocarbon-bearing zone 145 by creating additional paths 305. The injection fluid can also access a greater surface area or volume throughout the formation 140. Although the dilation or breakup of interbedded mudstones or shales is advantageous to speeding up the conditioning process, certain combinations of thickness of the hydrocarbon-bearing zone and permeability of the sand and mudstone layers may be such as to not require the interbedded mudstones or shale to be dilated or broken up to achieve conditioning in a reasonable amount of time. In any of the embodiments above or elsewhere herein, the rate at which the injection fluid is injected into the hydrocarbon-bearing zone 145 is dependent on the size, thickness, permeability, porosity, number and spacing of wells, and depth of the zone 145 to be conditioned. For example, the injection fluid can be injected into the hydrocarbon-bearing zone 145 at a rate of from about 50 barrels per day per well to about 5,000 barrels per day per well.In any of the embodiments above or elsewhere herein, the injection fluid can be injected at different depths within the formation 140 to access the hydrocarbon-bearing zone 145 therein. As mentioned above, the formation 140 can include embedded shale or mudstone layers that create baffles that

prevent flow or that surround or isolate one or more hydrocarbon-bearing zones 145 within the formation 140. The injection fluid can be used to create multiple fractures at different depths, i.e. both above and below the shale or mudstone layers to access those one or more hydrocarbon-bearing zones 145 within the formation 140. The injection fluid can also be used to create multiple fractures at different depths to increase the permeability throughout the formation 140 so the overburden 130 can be supported and overburden stress relieved more quickly. In any of the embodiments above or elsewhere herein, the injection fluid can be injected at different depths using a perforated lining or casing where certain perforations are blocked or closed at a first depth to prevent flow therethrough, allowing the injection fluid to flow through other perforations at a second depth. In another embodiment, the injection fluid can be injected through a perforated lining or casing into the zone 145 at a first depth of a vertical wellbore or first location of a horizontal wellbore, and the perforated lining or casing can then be lowered or raised to a second depth or second location where the injection fluid can be injected into the zone 145. In yet another embodiment, a tubular or work string (not shown) can be used to emit the injection fluid at variable depths by raising and lowering the tubular or work string at the surface. In yet another embodiment, two or more injection wellbores 110 at different heights could be used to create fractures in the formation 140. In general, this would remove the problem of trying to create multiple fractures from a single wellbore. Considering the injection fluid in more detail, the injection fluid is primarily water or brine during the conditioning phase. In any of the embodiments above or elsewhere herein, the injection fluid can include water and/or one or more agents that may aid in the conditioning of the formation or in disaggregating the shales or mudstones or the production of the slurry. Suitable agents may include but are not limited to those which increase the viscosity of the injected water or chemically react with the shales or mudstones to hasten their disagregation. In any of the embodiments above or elsewhere herein, the injection fluid can include air or other non-condensable gas, such as nitrogen for example. The ex-solution of the gas from the water can help dilate and fluidize the hydrocarbon-bearing zones 145 within the formation 140 as the solids are displaced into the lower pressure region near the production wellbore 120 where the gas could evolve from the water. In addition, the gas can help reduce the pressure

drop required to lift the solids to the surface by decreasing the solids concentration and overall density of the slurry stream in the wellbore. The gas can also help maintain higher pressure near the production wellbore 120 which would minimize the chance of the overburden 130 collapsing.

Transition Phase Once the stress from the overburden 130 is relieved and the hydrocarbon-bearing zone is conditioned, a pressure differential or pressure gradient is created between the injection wellbore 110 and the production wellbore 120. The developing or varying pressure differential between adjacent wells will cause water or brine to flow in the formation which will create fluid drag forces on solids in the formation 140. Once the pressure gradient in a given portion of the formation near the production wellbore 120 has increased to the point where it overcomes the friction holding the sand in place, the heavy oil, formation solids, and water will move or flow towards the production well. Therefore, this pressure differential moves or flows the formation 140 (sand, heavy oil, and water) toward the production wellbore 120. The flow or movement of the hydrocarbon-bearing zone 145 toward the production wellbore 120 can be referred to as “formation displacement.” It has been observed that the fluids in the hydrocarbon-bearing zone 145 (e.g., heavy oil and water) tend to flow relative to the solids and in the direction of the pressure gradient. The relative motion between the fluid and the solids creates a viscous drag (“drag force”), described by Darcy's law, on the solids tending to pull the solids towards the production well 120. This drag force is resisted, however, by the friction holding the solids in place (“frictional force”). Relieving or nearly relieving the overburden stress greatly reduces this friction, but the weight of the sand within the hydrocarbon-bearing zone and a small amount of residual overburden stress lead to a finite friction holding the sand in place. When the pressure gradient is high enough that the viscous drag force exceeds the frictional force holding the solids in place, the heavy oil, water, and solids will move in the direction of the low pressure areas of the reservoir (e.g. the producing wells). One method to develop this pressure gradient required to displace or mobilize the formation is to continue to inject fluid into the injection wellbore as was done during conditioning, but to reverse flow in the production wells and produce water rather than inject it as was done during conditioning. The

flow of water into the production well will set up a pressure gradient near the producing wells and when the pressure gradient is sufficiently large near the production wellbores a heavy oil, water, and solids slurry will start to be produced. As production continues, a pressure gradient will develop away from the production wellbores as a low pressure front propagates from the production wellbore towards the injection wellbore. As such, the zone of formation displacement will grow outward from the producing wells towards the injection wells as the pressure gradient is varied. When the zone where the pressure gradient is sufficient to cause formation displacement to occur reaches the injection wells, re-injection of cleaned sand and water slurry will be commenced. The length of time of this “transition period” from the onset of slurry production to the start of cleaned slurry re-injection will be dependent on slurry production rates, water injection rates, how the pressure gradient is varied, well spacing, and the effective permeability of the formation to the injected fluid(s). In addition to producing fluid from the production wells while continuing fluid injection in the injection wells, a pressure gradient may be developed by increasing the rate or pressure at the injection wells above those rates or pressures used during conditioning while producing some fluid (and eventually slurry) from the production wells. The relative rates or pressures of injection and production can be tailored to allow for the necessary pressure gradients to be developed while minimizing development of very low pressures around the production wellbores that could cause problems with slurry production into the wellbore. In any of the embodiments above or elsewhere herein, a water jetting technique can be used to emit the injection fluid into the formation 140. Preferably, the water jetting is a short, transitional step and used intermittently or for short periods of time. The water jetting technique can be performed through the injection wellbore 110 or the production wellbore 120 or both. In one or more embodiments, the water jetting is done through the production wellbore 120 after the formation 140 is conditioned to fluidize the sand and clay and create a slurry proximal to the production wellbore 120 opening allowing the slurry to be produced through the production wellbore 120. In addition, water jetting through the production wellbore 120 can remove any hard rock fragments that are too big to flow up the production wellbore 120 with the slurry. In addition to fluidizing a portion of the hydrocarbon-bearing zone proximal to the production wellbore,

water jetting may be used to further break-up or disaggregate shale or mudstone layers proximal to the wellbore to prevent them from impeding the flow of slurry toward the production well. During the production process, the movement or displacement of the formation towards the production well may allow the build-up of shale or mudstone near the production wellbore such that the flow of slurry into the production wellbore is impeded or the pressure gradient needed to move the formation increases beyond the pressure gradient that can be maintained. In such cases, additional water jetting in the production wellbore could be used to further break-up or disaggregate those shales or mudstones proximal to the production well and allow for them to be produced thereby allowing for unimpeded slurry flow into the production wellbore.

Production Phase: As discussed above, the hydrocarbon-bearing solids will move toward the production wellbore 120 provided the applied pressure gradient is large enough to overcome the frictional force holding the solids in place. The frictional force is proportional to the stress of the overburden 130 at the top of the hydrocarbon-bearing zone that is not balanced by the fluid pressure in the zone plus the buoyant weight of the solids within the hydrocarbon-bearing zone. In addition, there is some additional friction due to shearing forces as the displacing formation converges on the producing well and some additional friction at the base of the hydrocarbon-zone due to the viscosity of the heavy oil. Both of these forces in general will be smaller than the residual overburden and buoyant weight frictional forces.Furthermore, minimizing the stress applied to the solids by the overburden 130 minimizes both the pressure differential needed to move the solids and the injection rate needed to create the required pressure gradient. In addition, since the pressure gradient needed to displace the formation does not depend on the fluid viscosity (except slightly at the base) or on the permeability of the solids, as it does in conventional techniques of oil recovery, the high viscosity of the heavy oil or low relative permeability of the injection fluids does not increase the resistance to flow. As such zones within the hydrocarbon bearing zone that may have lower or high permeability or lower or higher water or oil saturation (and therefore

variations in fluid mobility in the zones) do not lead to a difference in slurry production from those zones as in conventional oil recovery processes.As mentioned above, the slurry for injection into the formation 140 contains the hydrocarbon-free or nearly hydrocarbon-free solids and recovered water from the recovery unit 170 and is recycled to the injection wellbores 110 via recycle stream 190. The solids, water, or mixture of the solids and water is then injected into the hydrocarbon-bearing zone via stream 150. Preferably, the injected slurry containing the recovered and recycled solids, water, or mixture of solids and water (i.e “re-injected slurry”) can include from about 35% to about 65% percent by weight of water, and from 65% to about 35% percent by weight of solids. In one or more embodiments, the injection fluid containing the recovered and recycled solids, water, or mixture of the solids and water can include of from about 40% to about 55% percent by weight of water, and of from 60% to about 45% percent by weight of solids. FIG. 4A is a schematic illustration to show the fluid dynamics within the formation 140 during an early production phase. Once the pore pressure (represented by arrows 410) is essentially equal to the overburden load (represented by arrows 420), a pore pressure gradient is developed across the formation by continuing to inject water into the injection wellbore 110 and produce slurry from the production wellbore 120. When the pressure gradient (fluid drag force) exceeds the frictional force holding the formation in place, the solids (represented by arrows 430) within the hydrocarbon bearing zone 145 will start to move toward the production wellbore 120, and a heavy oil-sand-water slurry will start to be produced through the production wellbore 120. FIG. 4B is a schematic illustration showing the re-injected slurry from the injection wellbore 110, solids 430 displacement toward the production wellbore 120, and production through the production wellbore 120. Once the pressure differential across the entire hydrocarbon-bearing zone 145 has exceeded the frictional force holding the solids in place, the solids 430 pull away from the injection wellbore 110 creating one or more voids 440. The re-injected slurry emitted from the injection wellbore 110 fills the voids 440 left by the displaced solids 430 and supplies the water needed to continue the displacement of the solids 430 toward the production wellbore 120 so additional oil-sand-water slurry can be produced through the production wellbore 120. Accordingly,

the re-injected slurry serves not only to dispose of the solids 430 removed from the hydrocarbon-bearing zone 145 but more importantly, maintains the integrity of the hydrocarbon-bearing zone 145. The solids within the re-injected slurry also suppress the tendency of the injection fluid to bypass over the top of the in situ hydrocarbon-bearing solids. Moreover, the re-injected solids will move more slowly once they enter the hydrocarbon-bearing zone if the permeability to the moving fluids is increased. This can have consequences for the optimal nature of the injected material. The permeability to water will typically be lower in the in-situ hydrocarbon-bearing solids than it would be in the same solids with the heavy oil removed. Hence, if the same solids are slurried with the water and used as the injection fluid, the in-situ hydrocarbon-bearing solids will tend to move faster in the hydrocarbon-bearing zone 145 than the reinjected solids. This can open voids in the hydrocarbon-bearing zone 145 with undesirable consequences. Therefore, it can be beneficial to add different materials to the reinjected solids to reduce the permeability to water. Optimally, this would be done in a manner so as to render the critical velocity of the mixed injected solids as it is in the in-situ hydrocarbon-bearing solids. In the hydrocarbon-bearing zone before slurry production begins, the clay, mud, and/or fine solid particles are generally concentrated in shale or mudstone layers. As such the overall absolute permeability in the horizontal direction of the formation is often dominated by the higher permeability sand layers. In some circumstances, the amount of this clay, mud, and/or fine solids could be such that when the hydrocarbon-bearing zone is completely disaggregated by flowing as a slurry up the production well and through the heavy oil removal process, this clay, mud, and/or fine particles become more evenly disseminated in the solids that are to be reinjected with recovered water into the injection wells. The overall absolute permeability of this material once it is reinjected may be significantly lower than the original hydrocarbon zone due to the dissemination of the clay, mud, or fine solids throughout the material. As such, in these circumstances the addition of additional materials to reduce the effective permeability of the reinjected material may be significantly lessened when the percentage of clays, mud, or fine solids is sufficiently high in the original hydrocarbon-bearing zone.It may also be advantageous to use one or more fluid/slurry injection techniques to locally (either spatially or temporally) increase the pressure

gradient. The term “pulse” or “pulsing” refers to variations or fluctuations in fluid or slurry injection or production rate or pressure. Such fluctuations can increase the pressure gradient locally to above the threshold for displacing the sand.

Stage 1 of SHORE process – “reservoir conditioning” via cold water injection into and pressurization of the reservoir the through

horizontal fractures

Stage 3 of SHORE process – steady state sand displacement with bitumen slurry production and cleaned slurry reinjection (after completion of Stage 2 “startup transition” with slurry production but water only injection)

SHORE process – steady state sand flow – map view of shape of moving sand lobes in 5-spot injector-producer pattern (from numerical model – red are sand velocity vectors, blue-green shading – shear strain magnitude)

Surface Facilities

Tailings processing Water reclamation

Bitumen extractio

Overburn

Schematic of SHORE closed loop process including surface

processing of the bitumen ore

Well ProductionW

ell I

njec

tion

Physics of the SHORE processThe key underlying concept of the SHORE process is that by relieving nearly all of the vertical, overburden stress on the reservoir sand, it takes only a moderate pressure gradient developed from water flow through the sands between injector and producer to move the sand towards the production well and then produce the slurry to the surface. The force balance can be written as:

dp/dx > bg*tan + (2v’*tan + 2C/h

Where h=interval thickness, =friction angle of the sand, C=cohesion of the sand, b= buoyant density of the sand pack, dp/dx=pressure gradient induced by fluid percolation through the sand, and v’ is the vertical effective stress on the sand.Thus, the pressure gradient to move the sand decomposes into the force to overcome the intergranular friction holding the sand in place due to the buoyant weight of the sand, due to the effective overburden stress on the sand, and due to the cohesion of the sand. For reservoirs about 10m thick with average mechanical properties found in the McMurray formation sands of Athabasca, the far right term in the above equation goes to zero, the middle term can be 2-15 kPa/m with less than 50kPa stress on the sand and the left term is 5-7 kPa /m due to the buoyant weight of the sand. Thus, only 10-20kPa/m of pressure gradient due to percolation of water through the oil sand ‒ as the bitumen is a nearly solid, immobile phase at reservoir temperatures is‒ sufficient during the steady state flow portion of the process to drag the sand lobe towards the producing well.The moderate pressure gradients are important to the process from both a slurry production and overburden stability perspective. Our modeling shows that the combination of water dilution and gas lift allows this dense produced slurry (generally 50-55% sand by volume at the wellbore) to be lifted to the surface for the range of well depths at which this process is effective. In addition, our modeling shows that 10-20kPa/m at 120-180m well spacing should not generate excessive deformation of the overburden from a cap rock integrityperspective.The “reservoir conditioning” phase of the process not only relieves the overburden stress so that the friction from the vertical effective stress is small, but the increase in permeability to water during

this conditioning is important to developing sufficient permeability to water in the sands to aid the sand displacement process. As shown in Yale et al. (2010), volumetric dilation of just 3% of an oil sand with a 1mD initial permeability to water can increase the permeability to water by a factor of 50. Our modeling shows that dilations on the order of 2-4% are likely for this process.In map view , the process produces petal shaped “lobes” of moving or displaced sand. The in situ bitumen sand within the lobe moves towards the producer and the re-injected cleaned sand fills in behind it. Laboratory and modeling results suggest that sweep efficiencies in excess of 50% are possible during primary production. As will be discussed later, the evolution of stresses within the moving sand lobes and in the pillars of non-moving sand are key to the why this process is effective in achieving high recovery. This evolution of stress off of the moving sand is also critical to allowing the process to operate at moderate pressure gradients during steady state sand displacement.

Experimental validationThe dramatic paradigm shift that the SHORE process represents over standard in situ recovery process necessitates significant validation of the process both experimentally and numerically before the field pilots can be used to assess the feasibility of the process. The process was first validated at the 25cm scale (Herbolzheimer and Chaikin, 1998) in a Lucite pressure vessel on loose sands with little to no effective stress on them. This simulated the process from a post-conditioning state. Large, symmetric lobes developed between the injector and producer and key physics of the process were validated.A steel pressure vessel was then constructed to test the conditioning through startup to steady state under stress conditions likely in real reservoirs. As shown in figure 6, lobes of re-injected sand were seen with high sweep of the in situ sand in the 20cm diameter by 2 cm thick sand pack that we similar to Herbolzheimer and Chaikin’s (1998) results. Although both sets of experiments validated the key physics of the process, determining how the process would scale up, what would be the impact of realistic heterogeneity, and collecting a dense enough array of data to validate numerical models of the process suggested another scale of experiments would be appropriate.a large-scale experimental apparatus was constructed and commissioned to test the SHORE process in 2m diameter by 5-20 cm

thick sand packs under full reservoir conditions (dubbed LARGE for Large-scale Apparatus for Reservoir and Geomechanics Experiments). Over 450 sensors on top, bottom, and within the sand pack have provided detailed insight into the feasibility and operation of this process. Stress, pressure, acoustic, flow, temperature, and resistivity sensors allow us not only to assess pressure gradients and effective stresses throughout the sand pack during the process, but also the real-time imaging (through conductivity contrasts) of the advancement, size, and shape of the lobes ofre-injected sand .As the bitumen in the SHORE process does not move relative to the sand, the process is controlled by the mobility of fluid movement through the sand pack. As such, all but a couple of small scale tests were done with a single phase fluid simulating the mobility of water in a bitumen-saturated sand pack (and thus accurately simulating the fluid and sand flow physics). However, the scale of the LARGE apparatus has afforded us the opportunity to test the process with a heavy oil-saturated sand pack. Although all the 20cm steel vessel and 2m LARGE tests have been done by compacting the sand to geologic conditions (glacial loading at depth followed by glacial unloading and uplift) and then conditioned through fluid injection to overburden stress, only the heavy oil-saturated tests have afforded us the ability to create a large horizontal fracture in the vessel and condition the sand pack from this fracture exactly as it would be in the field.It is also only in the LARGE vessel that the density of pressure and stress measurements have allowed us to fully understand the complexity of evolution of stresses and pressures in the reservoir interval during the startup transition into steady state sand flow. The LARGE vessel has also afforded us the opportunity to test the process with a simulated reservoir overburden. The polypropylene “spacer” that separates the sand pack from the hydraulic fluid that applies stress to the sand pack has been designed to mimic the deformation that a real reservoir overburden would undergo and more important how that deformation would impact the distribution and evolution of stresses on the sand pack. The similarity in sweep patterns and lab results between the 20cm and 200cm tests give us confidence that the process can be scaled to realistic well spacings (100-150m).

Integration of reservoir flow with well production-reinjection, surface processing, and caprock integrity Although the most novel aspect of this process is the massive sand displacement in the reservoir, the close-loop nature of the process requires a strong integration of the well, surface, and caprock integrity. Although the surface processing of the ore to strip off the bitumen, process the bitumen-water-fines froth, and process the solid tails is very similar to current aqueous surface mining processes, several key differences have been investigated to ensure a fully integrated, closed loop process will work.A key aspect of the surface processing-reservoir integration is the need for the processed tails to have a permeability within a range that allows the re-injected sand to move through the reservoir easily behind the moving bitumen sand. We have found that for a reasonable range of bitumen reservoirs and tailings processing schemes, it should be possible to inject most if not all of the tails into the reservoir interval and still achieve the reservoir sand flow objectives.We have also found that because of how the process “conditions” a reservoir interval, the “ore” that the SHORE process brings to surface to be processed may have different characteristics than the average ore that needs to be processed at most surface mining projects. As such, total water use and the ability to use brackish water from the subsurface may be quite different for this process than most current mining practices We have also found that wellbore flow needs and caprock integrity needs are surprisingly congruent. The range of well spacings, depths, and flow rates for which water dilution and gas lift work best for the process (relative to the pressures available after reservoir sand flow) are also those that work best for cap rock integrity.

Impact of geologic heterogeneityThere is a significant decoupling between the effectiveness of the SHORE process and geologic heterogeneity as compared to other “flow through porous media” based recovery methods. Reservoir zones with significant interbedded mudstones face recovery challenges with steam and solvent based processes as the mudstones serve as baffles if not barriers to steam chamber growth, bitumen flow to the producer, and/or solvent contact with a large reservoir volume. Although a significant fraction of mudstones can slow down the conditioning process, the conditioning process itself serves to break up mudstone barriers and create pathways through the mudstones. Figures 10 a-d show a numerical modeling of the conditioning process for a potential target reservoir which has a high degree of geologic heterogeneity. Although it takes over 12 months of water injection to reach a fully conditioned state deemed suitable to start sand flow in this case, the uniformity of the pressure field post-conditioning suggests a robustness of the process even with a significant level of interbedded mudstones. Models of lesser degrees of geologic heterogeneity suggest those reservoirs could be conditioned in 3-10 months.In addition, modeling shows that once the reservoir is conditioned, a fair number of interbedded mudstones will flow with the sand lenses towards the producer well and be produced up the well with the bitumen sands. Figures 11a-d show time slices of a reservoir model with 10% mudstone layers by volume and the small impact they have on the overall production. Modeling and laboratory work also show that there are large shearing forces within the moving sand bed as it converges on a production well. As such, we observe significant shearing of both sand and mudstone bodies into small enough pieces to be produced up the wellbore.More work is needed to fully assess the impact of heterogeneity, but the physics of the process and modeling and lab tests to date strongly suggest that reservoirs not amenable to a process like SAGD or those whose recovery may be significantly degraded due to their heterogeneity may have much more success with a process like SHORE.

Benefits of the SHORE processThis paper is meant as an overview of a promising new process for the in situ recovery of hydrocarbons from shallow bitumen resources. Our work to date has validated, from a numerical and laboratory perspective, that this novel process has significant technical potential to recover a wider range of bitumen resources than other current in situ technologies. We have found that the process is likely to operate with significantly smaller CO2 footprint than thermal recovery processes, with complete disposal of tailings back into the subsurface, and has the potential for near zero fresh water use. Recovery factors are likely to be in excess of 50% with bitumen production rates per well in the range of 1000-2000 bbls/day. Well spacings of 6-8 acres are similar to other in situ processes such that the high well production rates translate to draining of any given reservoir volume in 2-3 years.

Due to the physics of the SHORE process and its non-thermal nature, reservoirs down to 5m thickness should be accessible with this technology. Work to date also has shown the process has good potential to work in reservoirs with higher levels of interbedded mudstones or other geologic heterogeneity than other in situ recovery methods. Both these points along with observations of recovery factors in excess of 50% suggest that use of this technology may increase the bitumen recoverable from many leases over that which would be recoverable with many current thermal or solvent in situ processes.

The process is still in the early stages of development and faces many hurdles before its full technical and commercial potential at the field scale is known. However the process might eventually be shown to be an alternative in situ recovery method that has significant resource access and environmental benefits over standard thermal processes.

ReferencesBoone, T. J., Wattenbarger, C., Clingman, S., and Dickson, J. (2011) An

integrated technology development plan for solvent-based recovery of heavy oil, presented at SPE Heavy Oil Conference and Exhibition, Kuwait City, December 2011, SPE 150706

Boone, T.J. (1995) Exploiting high rate injection and fracturing to improve areal thermal conformance in cyclic steam operations, presented at SPE International Thermal Operations and Heavy Oil Symposium, SPE 79010

Butler, R.M. (1994) Steam-assisted gravity drainage: concept, development, performance, and future, Jour. Cand. Pet. Tech., vol. 33, p.44

Coates, R.M., LILLICO, D.A., LONDON, M.J., SAWATZKY, R.P., TREMBLAY, B.R.,(2002) “Tracking cold production footprints”, 53rd Annual Petrol. Soc. CIM Technical Meeting Proc. (Paper #2002-086)

Dusseault, M.B., LIAN, C.X., MA, Y., GU, W., XU, B., (2002) “CHOPS in Jilin Province, China,” SPE 79032Dusseault, M.B., (2006) “Sequencing technologies to maximize recovery”,

57th Annual Petrol. Soc. CIM Technical Meeting Proc. (Paper #2006-135)

Dusseault, M.B. (2007) “Cold Heavy-oil Production with Sand”, in Petroleum Engineering Handbook VI: Emerging and Peripheral Technologies, ed. H. R. Warner Jr., Society of Petroleum Engineers, Dallas

Fair, D. R., Trudell, C.L., Boone, T.J., Scott, G. R., and Spiers, B.C. (2008) Cold Lake heavy oil development- A success story in technology application, presented at the International Petroleum Technology Conference, Kuala Lumpur, December 2008, IPTC 12361.

Gronseth, J.M. (1993) Borehole mining process for recovery for petroleum from unconsolidated heavy oil formations, US Patent 5249844

Al-Sikaiti, S.H., and J. Regtien, 2008, “Challenging Conventional Wisdom, Waterflooding Experience on Heavy Oil Fields in Southern Oman,”

presented at the World Heavy Oil Congress 2008, Edmonton, Alberta, Canada, March 10–12.

Bethea, R.M., et al., 1983, “Effect of Groundwater and pH Modification on Quality of Leachate from Simulated In-Situ Retorted Utah Oil Shale,” Society of Petroleum Engineers Journal, SPE 10628 (Oct.):809−819.

BLM, 2008, Proposed Oil Shale and Tar Sands Resource Management Plan Amendments to Address Land Use Allocations in Colorado, Utah, and Wyoming and Final Programmatic Environmental Impact Statement, Bureau of Land Management, FES 08-32, September. Available at http://ostseis.anl.gov/documents/dpeis/index.cf m , accessed August 7, 2008.

Brost, D.F., 2002, "Water Quality Monitoring at the Kern River Field," presented at the 2002 Ground Water Protection Council Produced Water Conference, Colorado Springs, CO,Oct. 16-17. Available at http://www.gwpc.org/meetings/meetings_special/PW02/Papers/Dale_Brost_PWC2002.pd